Mycoplasma genitalium is a newly recognized cause of urethritis in men and possibly cervicitis, endometritis, pelvic inflammatory disease, involuntary sterility, and preterm birth in women. Despite the induction of an inflammatory exudate and specific antibodies during infection, this organism persists for months, if not years, in humans. I hypothesize that this persistence is associated with ability of M. genitalium to evade the immune response mediated by antigenic variation in two of its surface proteins, MgpB and MgpC, located in the complex terminally located attachment organelle. Despite its extremely limited genome, the smallest of any self-replicating cellular organism, 4% of the M. genitalium chromosome is devoted to partial incomplete copies of two genes, mgpB and mgpC, termed MgPars (for M. genitalium partial repeats). The sequence heterogeneity within mgpBC expression site and the homology with the MgPar[unreadable]s prompted the hypothesis that recombination between mgpB/mgpC and the MgPar sites results in antigenic variation in the resulting MgpB/MgpC proteins, thereby enabling this organism to evade the immune response and persist. In preliminary studies, we show that mgpB and mgpC are heterogeneous in the type strain, G37, cultured in vitro, and among M. genitalium strains detected in cervical specimens from persistently infected women. We also show that M. genitalium induces cervical and humoral antibodies in infected women. These findings provide powerful preliminary data for the studies proposed: (1) to evaluate the heterogeneity and homology of the mgpB, mgpC, and MgPar sequences within G37, the type strain of M. genitalium, to assess whether this heterogeneity is generated by gene conversion or reciprocal homologous recombination, and to determine the frequency, and effect of recombination genes, on spontaneous recombination in vitro; (2) to determine the surface exposure and cellular location of MgpB and MgpC epitopes, the biologic activity of monospecific antibodies directed against MgpB and MgpC, and the ability of these antibodies to select for antigenic variants of these proteins; and (3) to characterize the effect of antibodies to specific MgpB and MgpC epitopes on the evolution of mgpB and mgpC variability during persistent infection in women. These studies will lead to a broader understanding of the pathogenic mechanisms employed by this genomically-challenged organism, will reveal possible immune mechanisms leading to clearance of antigenic variants, and will inform future studies assessing possible candidates for future vaccines.